Official journal website: & Reptile Conservation amphibian-reptile-conservation.org 10(1) [General Section]: 7–19 (e114).

Epidemiological surveillance and amphibian assemblage status at the Estación Experimental de San Lorenzo, Sierra Nevada de Santa Marta, Colombia

1,*Luis Alberto Rueda-Solano, 2Sandra V. Flechas, 1María Galvis-Aparicio, 1Andres A. Rocha-Usuga, 3Edgar Javier Rincón Barón, 4Borish Cuadrado-Peña, and 5Rebeca Franke-Ante

1Facultad de Ciencias Básicas, Universidad del Magdalena, Santa Marta, COLOMBIA 2Departamento de Ciencias Biológicas, Universidad de los Andes, Bogotá, COLOMBIA 3Instituto de Biología, Universidad de Antioquia, Medellín, COLOMBIA 4Parque Nacional Natural Sierra Nevada de Santa Marta, COLOMBIA 5Parques Nacionales Naturales de Colombia, Territorial Caribe, Santa Marta, COLOMBIA

Abstract.—Amphibian population declines and extinctions have occurred in conserved sites or protected areas far from anthropogenic activities as a result of emerging infectious diseases such as chytridiomycosis. Regular epidemiological surveillance, monitoring of key species, and the implementation of biosecurity protocols are fundamental actions for the in-situ conservation of amphibian fauna. Since 2008 biosecurity protocols have been implemented for all personnel that enter the Estación Experimental de San Lorenzo, a partly mountainous protected and conserved area of the Sierra Nevada de Santa Marta with a high diversity of endemic and endangered . Semiannual disease screenings of amphibians were carried out, as well as an amphibian inventory and a survey of species of the . To-date no mass mortality events have been reported and Bd has not been detected. Nevertheless, some individuals of Ikakogi tayrona and megalops showed symptoms of disease, the latter of which included individuals affected with skin tumors. Deformities in individuals of Atelopus were also observed. The implementation of epidemiological surveillance, monitoring of key amphibian species, and biosecurity protocols are important strategies for the conservation management of the endemic amphibians within the protected area of the Sierra Nevada of Santa Marta.

Key words. Anura, Atelopus, Pristimantis, tumors, chytridiomycosis, disease screening, mortality events, health, dis- ease, Batrachochytrium dendrobatidis, Bd

Citation: Rueda-Solano LA, Flechas SV, Galvis-Aparicio M, Rocha-Usuga AA, Rincón-Barón EJ,Cuadrado-Peña B, Franke-Ante R. 2016. Epide- miological surveillance and amphibian assemblage status at the Estación Experimental de San Lorenzo, Sierra Nevada de Santa Marta, Colombia. Amphibian & Reptile Conservation 10(1) [General Section]: 7–19 (e114).

Copyright: © 2016 Rueda-Solano et al. This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommer- cialNoDerivatives 4.0 International License, which permits unrestricted use for non-commercial and education purposes only, in any medium, provided the original author and the official and authorized publication sources are recognized and properly credited. The official and authorized publication credit sources, which will be duly enforced, are as follows: official journal titleAmphibian & Reptile Conservation; official journal website .

Received: 02 March 2015; Accepted: 09 June 2015; Published: 31 March 2016

Academic Editor and Translation: Mayra Oyervides, The University of Texas-Pan American, UNITED STATES OF AMERICA

Introduction some level of threat (Baillie et al. 2010; IUCN 2014) and additionally about a quarter of amphibian species During the last decades amphibian population declines are classified as Data Deficient (DD, IUCN 2014), which and extinctions have been observed around the world, makes it more difficult to determine the actual status causing concern from academic, scientific, and of populations. Colombia harbors approximately 215 governmental entities (Gascon et al. 2007; Mendelson threatened amphibian species, which represent slightly et al. 2006; Stuart et al. 2008). Close to 41% of more than a fourth of its entire amphibian fauna (Acosta- amphibian species worldwide are categorized under Galvis 2014), making Colombia the country with the

Correspondence. Email: *[email protected] (Corresponding author)

Amphib. Reptile Conserv. 7 March 2016 | Volume 10 | Number 1 | e114 Rueda-Solano et al. greatest number of threatened amphibian species in the are still vulnerable to the threat imposed by Bd and other world (IUCN 2014). Nevertheless, some species have emerging diseases. For example, it is known that for the experienced serious declines while others remain stable. Parque Nacional Natural Gorgona, Bd has been present In addition entire lineages have been affected almost to for at least eight years (Flechas et al. 2012), however there the brink of extinction, as reported for the species of the is no evidence of declines. Additionally, amphibians are genus Atelopus (La Marca et al. 2005; Lötters 2007), vulnerable to diseases produced by aquatic pathogens, which in Colombia 76% (33 of 43) of the species are due to their dependency on aquatic environments. (Bosch categorized as Endangered and Critically Endangered 2003). For this reason authorities and administrators (IUCN 2014; Acosta-Galvis 2014). of protected areas in Colombia consider emerging Although continues to be the diseases (especially chytridiomycosis) a challenge to the main cause of population declines and extinctions protection of threatened amphibians. worldwide (IUCN 2014), it is puzzling that many species The implementation of recurrent epidemiological have disappeared in well conserved, remote areas such surveillance, monitoring of key amphibians species, and as primary forests in protected refuges (Crawford et biosecurity protocols, become fundamental to the in situ al. 2010; Crump et al. 1992; Lips et al. 2003; Pounds conservation of amphibian assemblages in protected ar- et al. 2006). We now have a better understanding of eas. This way, early alerts are generated and can be used the pathogenic microscopic fungus Batrachochytrium to implement and carry out the best management prac- dendrobatidis (Bd; Longcore et al. 1999), which causes tices in a timely manner, thus reducing Bd (or disease) the disease known as chytridiomycosis, one of the outbreaks and transmission. This paper shows the imple- leading factors behind mass mortality events previously mentation of a pilot program (the first program of its kind considered enigmatic. This fungus interferes with in the country) of these actions (surveillance, monitoring, the process of osmoregulation and affects electrolyte and protocols for amphibian species implemented for balance, which often leads to the death of susceptible disease control) set in a protected area, the Parque Na- individuals (Voyles et al. 2009). Furthermore, Bd seems cional Natural Sierra Nevada de Santa Marta, considered to inhibit the immune response of its hosts (Fites et al. one of the principle centers for amphibian endemism in 2013) which may present symptoms such as lethargy, Colombia (Lynch et. al. 1997). abnormal postures, and hyperemia (Berger et al. 2000; Daszak et al. 1999). Materials and Methods In Colombia, there are about 565 protected areas (RUNAP 2011), of these, 58 are administered by Study Area the Parques Nacionales Naturales de Colombia, and constitute approximately 11% of the continental territory The Sierra Nevada de Santa Marta (SNSM) was declared (PNNC 2015). These areas safeguard the country’s a reserve of the Biosphere in 1979 by United Nations Ed- biodiversity and represent a refuge for amphibian ucational Scientific and Cultural Organization (UNES- assemblages, including endemic species or species with CO). Situated within three departments of the Colombian narrow distributions, which may be susceptible to the loss Caribbean (Fig. 1, 2A.); 383,000 hectares belong to the and fragmentation of their habitat. However, these areas protected area (PNNC 2015). This area, comprises mul-

Fig. 1. Map of the Serrania de San Lorenzo, Sierra Nevada de Santa Marta, Colombia. Red square area highlights the Estación Experimental de San Lorenzo to 2,200 meters.

Amphib. Reptile Conserv. 8 March 2016 | Volume 10 | Number 1 | e114 Epidemiological surveillance and amphibian assemblage status

November, with a dry period between December through March (Tamaris-Turizo et al. 2007). The mean annual temperature is 12.8 °C, the mean annual precipitation is 2,446 mm and the relative humidity oscillates between 73–98% (Tamaris-Turizo and López-Salgado 2006).

Epidemiological Surveillance

Since 2008, population census have been performed employing the visual encounter survey (VES) meth- od (Rodda et al. 2001; Rueda et al. 2006; Heyer et al. 1994) in determining numbers of with clinical signs of diseases, present deformities and/or individuals found dead in the surrounding areas of the Estación Ex- perimental San Lorenzo. Epidemiological surveillance has been carried out through programmed visits every six months by researchers and biology students from the Universidad de Magdalena and through scheduled or nonscheduled visits by staff of the Parque Nacional Natural SNSM. We looked for anurans presenting leth- argy, macroscopic lesions, abnormal postures, hyper- emia, ulcers or the presence of fungi or other corporal anomaly, such as deformities. Data was collected for all individuals presenting clinical signs of a potential Bd in- fection. Swab samples from individuals were obtained Fig. 2. Serrania de San Lorenzo (B) Querbrada San Lorenzo following the protocol described by Hyatt et al. (2007). (A) Sierra Nevada de Santa Marta, Colombia. Photographs by were captured using fresh disposable nitrile gloves Luis Alberto Rueda Solano. and held individually in bags until the sample was taken. tiple ecosystems, including dry and wet tropical forests, Each was swabbed by running a cotton swab ten sub-Andean and Andean forests, moors, and zones with times over the ventral surface, the inner thigh area, and perpetual snow cover (PNNC 2015; ProSierra 2015). the plantar surface for a total of 50 strokes. Cotton swabs With approximately 17 species of amphibians, 12 of rep- were kept dry and then stored at -20 °C until processing tiles, 14 of birds and one mammal, all of them endemic to of future laboratory analysis. the area (PNNC 2015; ProSierra 2015), it is considered One single specimen of Pristimantis megalops with one of the greatest centers of endemism in the country clinical disease signs was analyzed using histological and one of the irreplaceable protected areas of the world methods. The samples of the skin areas affected with le- (Le Saout et al. 2013; Lynch et al. 1997). sions were fixed in FAA (Formalin alcohol-acetic) for The study site is the Estación Experimental de San 24–48 hr at 6 °C. Subsequently, dehydrated in graded se- Lorenzo (11° 6’ 41.61” N 74° 3’ 17.13” W), located in ries of alcohols (30, 40, 50, 60, 70, 80, 90, 95, and 100%) the Serranía de San Lorenzo, on the northwestern slope and two cleared steps in xylene for two hours, then em- of the SNSM, department of Magdalena, Colombia (Fig. bedded in Paraplast Plus (Mc Cormick®) for 12 hr at 55 2A) at 2,200 m and comprising an area of 400 ha. The sur- °C (Luna 1968; Suvarna et al. 2012). Transverse and lon- rounding vegetation is comprised of well conserved par- gitudinal sections were obtained with a rotary microtome tially mountainous primary and secondary Andean for- Leica ® model (RM2125) set to 4–5μm thick. These were est, which include tropical and subtropical rainforests of stained with hematoxylin-eosin for general descriptions the isomesothermic jungles (14 °C to 24 °C) (Hernández- and were previously stained using the Van Gieson tech- Camacho and Sánchez-Páez 1992). The site’s native flora nique with alcian blue and Gill III hematoxylin (changes includes woody species of Gustavia speciosa, Sloanea made by the authors) to demonstrate collagen in the con- sp. and some palm species of the genus Geonoma sp. and nective tissue. The sections were examined under a light Chamaedorea sp. (Cleef and Rangel 1984; Rangel and microscope Nikon Eclipse Ni-U® equipped with differ- Garzon 1995). Nevertheless, some hectares of non-native ential interference contrast (CDI). The photographs were vegetation are present. Coniferous forests with species obtained with DS-Fi2® Nikon digital camera using the of Pinus patula and Cupressus lusitanica introduced in NIS Elements of Nikon software version 3.07. The image the early 90s, may influence the amphibian assemblage processing was performed with Image-Pro Analyzer 6.3 that inhabit this area (Camero and Chamorro 1999). This program (Media Cybernetics). These analysis were car- sector presents one rainy season, between April through ried out in the biology laboratory of the Universidad de

Amphib. Reptile Conserv. 9 March 2016 | Volume 10 | Number 1 | e114 Rueda-Solano et al.

Antioquia (Medellin, Colombia) and photographs taken transmission of Bd, since 2008 park administrators from in the Laboratory of Biotechnology of the Sede de Inves- SNSM have been implementing a biosecurity proto- tigacion Universitaria (SIU-UdeA). col for all foreign and national personnel that enter the Furthermore, a single specimen of Ikakogi tayrona Estación Experimental de San Lorenzo. The protocol with clinical disease signs was analyzed using conven- consists of disinfection of field equipment (boots, nets, tional Polymerase Chain Reaction (PCR) to determine measuring devices) used as much by researchers as by Bd presence. DNA was extracted from swabs using Ge- tourists who visit the protected area (Phillott et al. 2010). neReleaser® (Bioventures Inc., Carlsbad, California, Nevertheless, as a preventive measure in situ, field USA). We used the primers developed by Annis et al. boots are washed with approximately 200 mL of a com- (2004) to amplify the ITS1-ITS2 region specifically in mercial sodium hypochlorite solution diluted in three li- B. dendrobatidis: Bd1a (5’-CAGTGTGCCATATGT- ters of water. CACG-3’) and Bd2a (5’-CATGGTTCATATCTGTC- CAG-3’). Amplifications were performed in an MJ Re- Amphibian Assemblage search Peltier Thermal Cycler (PTC-200), as follows: an initial two minute denaturation at 95 ºC followed by 35 To describe the amphibian assemblage from the San Lo- cycles of DNA amplification (i.e., 45 sec at 95 ºC, 45 sec renzo sector, an inventory was carried out during seven at 55 ºC, and one min at 72 ºC). A final extension at 72 surveys throughout the months of October and Novem- ºC for 10 min completed the amplifications. Each reac- ber of 2008, March of 2009, April and October of 2013, tion consisted of 0.5 μL of each primer (1 M), 3.0 μL of and April and November of 2014. No surveys were made doubly distilled DNA-free water, 6 μL of GoTaq® Green in the years 2010–2012. Each field trip had a duration Master Mix (1X; Promega), and 2 μL of the DNA extract. of four days, and the study site was surveyed for eight The amplified fragments were separated by electrophore- hours daily, employing the VES method in diurnal peri- sis through 1% agarose gels. These analysis were carried ods (9:00–12:00 and 15:00–17:00 hrs) and nocturnally out in the genetic laboratory of the Universidad de los (18:00–21:00 hrs). During the surveys, for each individ- Andes (Bogotá, Colombia). ual data associated with the habitat and time of day of the observation were registered. The number of research- Biosecurity Protocols ers varied among surveys (from two to seven), and thus each survey had different effort levels (between 16 and To date the presence of Bd has not been reported at Ser- 80 hours × person). To determine the relative abundance ranía de San Lorenzo. To prevent and reduce the risk of (RA) of different species in the sector, a classification of

Table 1. Epidemiological surveillance from the year 2008 to 2014 for each of the reported amphibians within the sector of the ex- perimental station of San Lorenzo, 2,200 m altitude, Sierra Nevada of Santa Marta, Northern Colombia. Very rare (VR); Rare (R); Common (C); Abundant (A); Very Abundant (VA). Year of Nº de- Nº diseased Disease Nº dead reported Species n RA Microhabitat Habit formed individuals type individuals sick indi- individuals vidual Terrestrial/ Cutaneous Atelopus laetissimus 128 VA Nocturnal 1 1 0 2013 Shrubs ulcers Terrestrial/ Diurnal/ Atelopus nahumae 10 R 0 — 1 0 2013 Shrubs Nocturnal Undeter- Ikakogi tayrona 26 C Shrubs Nocturnal 1 0 0 2008 mined Pristimantis delicatus 13 R Shrubs Nocturnal 0 — 0 0 — Pristimantis carmelitae 23 C Terrestrial Nocturnal 0 - 0 0 — Pristimantis cristinae 5 VR Shrubs Nocturnal 0 - 0 0 — Pristimantis insignitus 4 VR Terrestrial Nocturnal 0 - 0 0 — Fibropap- 2008; Diurnal/ Pristimantis megalops 477 VA Terrestrial 9 illoma 0 0 2013; Nocturnal (tumors) 2014 Pristimantis ruthveni 15 Rare Shrubs Nocturnal 0 — 0 0 — Pristimantis sanctaemartae 211 VA Shrubs Nocturnal 0 — 0 0 — Pristimantis tayrona 48 A Phytotelmata Nocturnal 0 — 0 0 — Pristimantis sp. nov. 1 29 C Shrubs Nocturnal 0 — 0 0 — Terrestrial/ Pristimantis sp. nov. 2 21 C Nocturnal 0 — 0 0 — Shrubs Shrubs/Phyto- Bolitoglossa savagei 44 A Nocturnal 0 — 0 0 — telmata

Amphib. Reptile Conserv. 10 March 2016 | Volume 10 | Number 1 | e114 Epidemiological surveillance and amphibian assemblage status

Fig. 3. Healthy individuals of Atelopus laetissimus (A) (Bufonidae); Atelopus nahumae (B) (Bufonidae); Pristimantis megalops (C) (); and Ikakogi tayrona (D) (Centrolenidae). Photographs by Luis Alberto Rueda Solano. very rare, rare, common, abundant and very abundant, and at night (18:00–00:00 hrs). For individuals of Atelo- according to the of individuals (ind.) recorded during all pus, SVL, weight, and sex have been systematically re- the surveys was established. Species were very rare, if it corded since 2013. Each individual was handled with a was observed equal to or less than nine ind.; rare, if it was new pair of gloves as part of the biosecurity protocols to observed between 10–20 ind.; common, if it was observed prevent the transmission of Bd. between 21–30 ind.; abundant if it was observed between 31–50 ind.; and very abundant, if it was observed over RESULTS 50 ind. in the total of all surveys. For identification of the species belonging to the genus Pristimantis we employed Epidemiological Surveillance the synopsis of Lynch and Carranza (1985) and for the remaining species we used the information provided by To date no mass mortality events have been reported, or the American Museum of Natural History from its online individuals with field evidence ofBd infections in the as- reference Amphibian Species of the World (Frost 2014). semblage of the 15 endemic species of amphibians at the San Lorenzo sector, corresponding to 1,375 ind. mostly Monitoring of Atelopus laetissimus and healthy individuals of Bufonidae, Craugastoridae, Cen- Atelopus nahumae trolenidae, and Plethodontidae (Table 1, Fig. 3). How- ever, 13 sick individuals of the species Ikakogi tayrona, Due to the importance represented by the species of At- Atelopus laetissimus, A. nahumae, and Pristimantis meg- elopus, for being one of the most affected genus for Bd alops were recorded (Table 1, Fig. 4). (La Marca et al. 2005; Lotters 2007), we monitored both The sick individual of Ikakogi tayrona presented species reported at this locality, A. laetissimus and A. na- symptoms similar to those of chytridiomycosis, such as humae (Ruiz-Carranza et al. 1994). These species were lethargy, pale skin, and hyperemia of the ventral ski